This is the third episode of a book club series on Peter Gärdenfors's book Conceptual Spaces. In this episode, we will discuss chapters 5 and 6, in which Gärdenfors explains how semantics and induction fit into his theory of conceptual spaces.
For this series, I'm joined by Koen Frolichcs, who was already my cohost for the books club series on Lee Child's Killing Floor. Koen and I are PhD students in the same lab.
Podcast links
Website: https://geni.us/bjks-pod
Twitter: https://geni.us/bjks-pod-twt
Koen's links
Google Scholar: https://geni.us/frolichs-scholar
Twitter: https://geni.us/frolichs-twt
Ben's links
Website: https://geni.us/bjks-web
Google Scholar: https://geni.us/bjks-scholar
Twitter: https://geni.us/bjks-twt
References
Gärdenfors, P. (2004). Conceptual spaces: The geometry of thought. MIT press.
Hohwy, J. (2013). The predictive mind. Oxford University Press.
https://en.wikipedia.org/wiki/Homo_economicus
-
[This is an automated transcript with many errors]
Benjamin James Kuper-Smith: [00:00:00] This is now the third part of our discussion of Peter GaN force's book Conceptual Spaces. Um, I'm again joined by con.
Koen Frolichs: Hello.
Benjamin James Kuper-Smith: And uh, today we'll be discussing chapters five and six, which deal with semantics and induction. I think Kun are my favorite topics.
Koen Frolichs: Yeah, exactly.
Benjamin James Kuper-Smith: Yeah. And, um, yeah, so last time actually, you know, we said like the, the looking forward to this part, we said the.
I think the semantics chapter was 60 pages long and the induction chapter was 30 pages long. Yeah. And I kind of wished it was the other way around, but I think having read them, I was pretty happy that it was this way around, not the other way around, because I kind of actually enjoyed the semantics one, despite my expectations.
Koen Frolichs: Okay. Very,
Benjamin James Kuper-Smith: uh, and I found the induction one largely confusing.
Koen Frolichs: Mm-hmm.
Benjamin James Kuper-Smith: Um, I, maybe I wasn't particularly focused per se, but Sure. Yeah, I, I didn't really take much from that, to be honest. [00:01:00] Yeah. I guess maybe it's, uh, to say it as at the beginning, I don't think this episode's gonna be particularly long. Uh, I don't think either of us has a huge amount to say here.
Um,
Koen Frolichs: Nope. I actually really
shot.
Benjamin James Kuper-Smith: Yeah. I don't,
Koen Frolichs: yeah. No. Me neither.
Benjamin James Kuper-Smith: Yeah. Uh, so may, this might be a thing where we just say. Yeah, we read them. Thank you for listening.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: Next week will be chapter seven and eight.
Koen Frolichs: Yeah. If you want some information, like do it yourself. Yeah. No, I mean, I kind of expected this and it was kind of as bad as I expected, and this is not like me shedding on the writing or anything.
It's just that I find this, you know, this is not my topic of choice.
Benjamin James Kuper-Smith: Yeah.
Koen Frolichs: And to read it in such an abstract. Manner, like, or like, you know, like, it's like, it's not semantics, right? It's like abstracting semantics into like something that you can like work with. Um, I just, I just found I find incredibly hard.
Um, you know, like I,
yeah.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: Yeah, I mean I think for me the main problem is just that I just don't know anything about [00:02:00] semantics. Yeah, I mean, I've heard of the topic, I've probably encountered it a few times in passing, but I just dunno anything about it. So I, you know, I read through the thing and there were a few things I found kind of interesting how like, you know now, like adjectives can be like.
Uh, parts on a dimension or something, and now it's can be these clusters in it and, uh, what was it? Clusters? Uh,
Koen Frolichs: yeah, I think so.
Benjamin James Kuper-Smith: Yeah. Or like multiple things like, you know, there were a few things in there that I thought were kind of interesting, but I just, I have no comparison or reference point basically when talking about semantics for me it's just like, this is more or less the first time I'm hearing about it in a detailed way.
Koen Frolichs: Yeah. Yeah.
Benjamin James Kuper-Smith: Yeah. And I just have nothing really to add other than Sounds kind of logical. It seems like the whole thing. And that's, you know, the way he puts it also is that it's still very much at the beginning.
Koen Frolichs: Mm-hmm.
Benjamin James Kuper-Smith: And I think his whole idea is more like. This is something that people think of this way, but I think it makes more sense to think of it this way and it kind of fits in.
Koen Frolichs: Yeah, yeah, yeah, yeah.
Benjamin James Kuper-Smith: But it's also very much an [00:03:00] early field, it seems to me. So it seems like a lot of stuff isn't worked out, but yeah. I just don't have much to add.
Koen Frolichs: He even said so in the, in the end, right. It's like this is like, this is me kind of throwing these ideas out there, I think at the end of chapter five.
Yeah. This is not like, it's not completely thought to it. It's a lot of work still. Um, I don't know if it has had worked since 2000
Benjamin James Kuper-Smith: probably,
Koen Frolichs: but past 20 years. Yeah, probably. Yeah. Um, but, but yeah, you're right. I guess what, what I found really difficult, what I found easier in the previous chapters was also to see it in this.
Framework, more like I felt like either we went back to the framework more, or I could myself put it back, you know, in this conceptual framework and now I felt like there were a lot of pages of, you know, semantics and then like a little bit like, you know, only like two or three, like when the figures came where you're like, ah, okay.
You know, as you just said, you put the clusters, whatever. And I found that really difficult that it makes it so dry. Um,
Benjamin James Kuper-Smith: yeah. Yeah. One, one small point I wrote down somewhere is [00:04:00] that I really felt like whenever he started using examples, I started being engaged again.
Koen Frolichs: Mm-hmm.
Benjamin James Kuper-Smith: Um, to some extent it feels like it did feel like he used quite a few, quite a lot actually, but
Koen Frolichs: Sure.
Benjamin James Kuper-Smith: Yeah. It's just every time I was interested I realized like, oh, it's, he's, because he's using examples here too. Show how something fits into something. And yeah, I mean, again, maybe for people who are really steeped into semantics, these, you know, they're like obvious topics to talk about. You don't need to explain them much.
Um, the same way, like if I were to read something about cooperation, I probably wouldn't want an example for everything. Every point made. Yeah, exactly. Because I've probably heard them a lot before, but
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: Yeah. In this case I hadn't. Um,
Koen Frolichs: yeah, you kind of need to be taken by the hand more, which, you know, he probably doesn't even have the space for.
Right. To, to
Benjamin James Kuper-Smith: that's the next thing then. It's obviously twice as long if he does that. Yeah. So, I dunno whether we want that either.
Koen Frolichs: Do you know, do you know a favorite example of, of chapter five or six that he gave? Or like, uh, an idea? Do you have one like that? Just, I just have one pop in my head. That's [00:05:00] why,
Benjamin James Kuper-Smith: uh, you can just say that one, then I'll have a look.
Koen Frolichs: What I thought was interesting, the me the metaphors, like the me mapping of metaphors, right? Yeah. How you like take the structure from an existing concept and then apply it to a new thing and that makes sense. I thought that was kind of an, an interesting idea. Right? But that kind of also, I guess that also fits back into this Cartesian way of looking at it kind of, right?
Like the top of the mountain, the top of the
Benjamin James Kuper-Smith: mm-hmm.
Koen Frolichs: Social hierarchy, stuff like that. Um. So, yeah, I guess that's, that's when I was kind of engaged against like, oh yeah, okay, yeah, this, I get it. Get right.
Benjamin James Kuper-Smith: Exactly. That's what I mean. That's the kind of stuff that I find interesting. And also, I think he also addresses this briefly, but at some point I did have this question of, you know, I think he says something like that the main metaphors are spatial or something like that.
Mm-hmm.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: Yeah. True.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: You know, again, like I came to this book through the space navigation literature. Yeah, yeah. And there's the same kind of question, like, do people. Uh, how should we say like, I guess like, [00:06:00] did grid sales kind of first and play sales and that kinda stuff first kind of, were they used for space and then you can use them for other stuff also?
Or is it this generic, generic thing that was from the beginning generic. Mm-hmm. Um, and I guess he kind of slightly elude to this also by saying like, spatial ones seem to be a bit more fundamental maybe, but
Koen Frolichs: Yeah, exactly. Yeah. I thought it was interesting as well, right? Yeah. It's like a, a, a little knot towards like.
We were spatial beings first and then the rest came kind of right. Or like,
Benjamin James Kuper-Smith: not that I guess because if you look at, yeah, but like, I mean, I guess if you look at, you know, take like, uh, the simplest organisms, the first thing they can do is move towards and away from just something, right?
Koen Frolichs: Yeah, yeah. You're right.
Benjamin James Kuper-Smith: Yeah. So I guess movement is, and space is kind of the, I think, I don't know, maybe there's, there's probably like, uh, there's probably some bacteria that do something else first, but, um,
Koen Frolichs: sure.
Benjamin James Kuper-Smith: I dunno. I. I think, uh, I don't know. One, the, the one point I, uh, wrote down about this one, about this [00:07:00] part is kind of one question I've had in the beginning.
I don't think I've, I think I've wanted to ask it before, but I don't think we got to it and that he kind of eludes to here also. That he cites, uh, Putnam, whoever that is, who has this example around his confusion or his lack of knowledge about the difference between elms and beaches. That he basically, you know, there's these two types of trees and he's like, I dunno, which is which, like for me it's all the same basically.
And he uses that as some sort of. I can't remember exactly, uh, but some example of why X or Y doesn't work or whatever. Yeah. And, uh, but the one thing that I found interesting here that relates to something I was interested in is just generally this question of, and I think this is probably a really interesting empirical question also.
It's just kind of like, how do you represent. No, let's, let's forget about semantics and that kinda stuff. For the, for now. Just go back to the general idea of the, uh, concepts made up of several dimensions. Like how do you represent something where, [00:08:00] or basically how do you represent something where you don't know what the, what a value on a dimension is.
For example, I dunno, let's say you, you, I, I say, Hey, there's this person, my friend Mike.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: And you know nothing about this person. I mean you, I don't know. Right. So you have all these dimensions, you have no information. Do you just go with a prototype? Do you just, do you actually not represent any other information than the name and the gender right now?
Or do you kind of, yeah, I was just, uh, so that's kind, I guess one question is kind of like, how is someone, something represented if you have very little information about it, but there's lots of variables that. You know, are relevant usually.
Koen Frolichs: Mm-hmm.
Benjamin James Kuper-Smith: And the other is basically like, how do you, 'cause you know, that's kind of with his elms and beaches is there's all sorts of differences.
He just doesn't know what they are basically.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: And the other kind of related point to this is that, yeah, I guess kind of when you have to make a dec well, not decision, when, [00:09:00] you know, when you, when you process some, let's say again a person or something, and there's just lots of variables and dimensions that aren't relevant.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: Are those process the entire time or do you kinda just lose the irrelevant ones? So depending on context, you just pick out the dimensions that seem most relevant. Um, that's
Koen Frolichs: interesting. Yeah.
Benjamin James Kuper-Smith: Do you know what I mean? Like you have in some sense this representation, like the, the, the complete representation of something.
Mm-hmm.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: Uh, that might, in your case, not even be complete because you lack information. But then also when you use it like. Is it like you have like this, this nucleus or something of, you know, you, you've kind of saved, let's say, using the computer metaphor, like you've saved lots of information about something and then you only like pick out the thing that's relevant, you pull out certain information.
Yeah, I dunno. That's just the one, the one thing in here that for like. That relates to something I'm actually really interested in knowing what it is.
Koen Frolichs: Yeah, yeah. No, it is. It is really interesting and I think there's [00:10:00] probably like several answers that make sense or that could be tested right, empirically.
Yeah.
Benjamin James Kuper-Smith: Right. I mean, I'm asking very general questions, so there's probably lots of different answers for it.
Koen Frolichs: Yeah. No, but it is, it is interesting, I think, and I think Peter, well, probably would say like, it's like context dependent, right? Like that's kind of,
Benjamin James Kuper-Smith: yeah, yeah.
Koen Frolichs: Like also like a, what dimensions you focus on and also like how you start representing this person, right?
Like let's say you would always use a prototype, but then even like if I think about like a friend of yours could be. Different prototypes. I do. I look at you as Ben, my colleague. Right. Then you're right. Mike is probably gonna be
Benjamin James Kuper-Smith: like, in which context do I mention this person?
Koen Frolichs: Exactly, yeah. And like Yeah, exactly.
And like, where are you in my context? Right. Are you like, you know, a, a very good friend or like an acquaintance, whatever, like, right. Yeah. So I guess that's. Yeah, that's probably it. That's, but
Benjamin James Kuper-Smith: I feel, feel like that is something that's super, like it also kind of relates I guess like when I'm interested in like how people make decisions in social context.
There is this kind of question of which information, you know, [00:11:00] this is the one thing I think about a fair bit or
Yeah.
Benjamin James Kuper-Smith: Want to, in the future more, what information do people use in a given situation? And then I guess the, the, the question here that I'm kind of asking in how it's represented. When this information has to be used is maybe more question of like, biological implementation.
And, um, yeah, I'm not entirely sure. I mean, I guess there's just a bunch of things. Um,
Koen Frolichs: yeah, yeah. It's a very broad, um, um, thing you brought, brought up now, but, but, but yeah. Yeah. No, but I think it's a, it's a very interesting question. Yeah. I mean, it's kind of like this example he gives with the, the color red, right?
Red book, red wine, red hair. Like that's, that's kind of like. I guess is it, it feels a little bit related to what you're saying, but I, I, I dunno how to like. I guess, yeah, that's just a context thing, I guess. It's not that, that important.
Benjamin James Kuper-Smith: Yeah, yeah.
Koen Frolichs: Like,
Benjamin James Kuper-Smith: yeah. I mean actually that, actually, actually that was a thing I found quite interesting, um, this kind of like how people differentiate colors.[00:12:00]
Like first you have black and white, then you add red or something. What was it? There was this kind of like order in which people basically, like at first you go, like, you just differentiate them based on. Brightness or something, and then you, you know, then you kind of, basically, you only need the smaller colors once the larger categories don't work.
Koen Frolichs: Yeah,
Benjamin James Kuper-Smith: yeah, yeah. Um, you know, like if you have, uh, what, what would, yeah, I can't remember exactly what the example would be now.
Koen Frolichs: Lilac or a pronou?
Benjamin James Kuper-Smith: Lilac, but yeah,
Koen Frolichs: lilac.
Benjamin James Kuper-Smith: Okay. And I think, I'm sure. But I just find it interesting that you have this like clear hierarchy in terms of like more or less in which order people start using certain words.
Yeah. Which, you know, makes complete sense, right? That you wouldn't, if you have a new species of animal, you wouldn't go like immediately down to
Koen Frolichs: Yeah. Yeah.
Benjamin James Kuper-Smith: Like this is precisely the color of thing you've said, like Yeah, it's a kind red bird kind of.
Koen Frolichs: Yeah, yeah, yeah, yeah. It's like a dark with wings.
Benjamin James Kuper-Smith: Exactly. And [00:13:00] red-ish.
Koen Frolichs: Redish. Yeah.
Benjamin James Kuper-Smith: Yeah. I mean, one, one related thing here that's kind of random, but uh, so there's this thing, I heard that basically until fairly recently there was the color orange. Didn't exist in Yeah, true German or English or whatever. I can't remember. But you know that the idea that basically orange is one of the few colors that's named after the actual fruit.
And since I think the fruit just didn't exist in, I don't know whether it's just like German or English countries or in Europe or something like that, until, yeah, a few hundred years ago. But that, for example, like in, uh, I can only think of German examples now. Um, the, the bird species I got, let me look up the English name for this.
Uh, so it's just called the European Robin.
Koen Frolichs: Oh,
Benjamin James Kuper-Smith: uh, in German is called, uh, actually we can also call it an English robin Red breast. Okay. Ah, yeah. Okay. Yeah, sure.
Koen Frolichs: Yeah, yeah.
Benjamin James Kuper-Smith: You know Ian in German, that's the Yeah. And the thing is, it's obviously not a red chest. It's orange. Truly, [00:14:00] I always thought
Koen Frolichs: they were red.
Okay.
Benjamin James Kuper-Smith: No, it's, it's orange and there's apparently quite a lot of things like that in English and German where people say it's red, but actually it's more orange. It's just they didn't really have the color then. So I wonder to some extent whether because red was really early in the hierarchy
Yeah.
Benjamin James Kuper-Smith: That the people listed.
So I wanted to what extent that depends on the language you use and the words for colors you have in them. Um,
Koen Frolichs: yeah. Makes sense.
Benjamin James Kuper-Smith: But anyway, I still like the, um. The general kind of, I thought it was interesting that, that it would be kind of so. Clear in a se in a way.
Koen Frolichs: A, actually, I, what I was thinking is that that's probably also why there's so much interest in semantics and stuff like that, because that's kind of, I'm not saying this is easy to study, but that like before like, you know, like neuroimaging techniques, whatever, right.
This was a way of really looking into cognition, right? Like what words do kids start like that? You had an example. Like kids. Yeah. The first called things dogs and not animals.
Benjamin James Kuper-Smith: Oh yeah. Yeah. I [00:15:00] think they had
Koen Frolichs: like, and almost like too broad of a category to understand initially or
Benjamin James Kuper-Smith: so understand. Yeah, that was, yeah, there was some legacy.
It's kind of interesting, wasn't it?
Koen Frolichs: I have to admit. Yeah, yeah. 'cause I always like shit on people and they're like, oh, um, um, um, language is so interesting. It's like, but um, no, but yeah, no, it is. Yeah. And actually like, you know, and it from a developmental perspective. Like, it's very, like, it is like a window into, into cognition.
Right. Which is
Benjamin James Kuper-Smith: right. And how they learn about rules and that kind of stuff. When, when children do grammatical errors that are actually correct, you know, like, uh, yeah, yeah. Fight it or something like that. Yeah. Um, that's, you know, it's completely correct reasoning, but it's just incorrect practically.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: Um, but. Yeah, it's funny. I mean, yeah, we, we, yeah. I always say language is so boring, but then when you read it about it, it's like, oh, this is really cool.
Koen Frolichs: Yeah. Yeah. No, it definitely, it definitely has some, some very interesting things. Yeah. And I would, I was thinking like, you know, this is kind of how, I'm not sure if this is complete true, but kind of like the, [00:16:00] the earlier studies into cognition, right?
That was, I think, a lot of LA language based. And I think it makes sense, right? Because it's, you know, it's a good way of studying it or like, you know, in like, at least like initial predictions.
Benjamin James Kuper-Smith: Definitely. Um, do you have anything else? I think maybe we should just stop whilst we're still positive about language.
Koen Frolichs: Exactly.
Benjamin James Kuper-Smith: Do you have, or do you have anything else there? Say about chapter five?
Koen Frolichs: No, I think that, no, I think that was it for chapter five. Um,
Benjamin James Kuper-Smith: okay, cool. Let's sit on chapter six then. No, I'm still gonna be kind, but I have to admit, I, I mean, I dunno, I read this today and maybe I just wasn't particularly structured or whatever, but I find it slightly found it slightly hard to
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: Follow exactly what he was trying to say and what the point of the whole chapter was. I mean, he did clearly say in the beginning what the point was, but I'm just not sure whether, yeah, I dunno. I, I just somehow thought this was gonna be a lot more interesting than I ended up finding it.
Koen Frolichs: Yeah, yeah, yeah, yeah.
This was like, because chapter five, I knew I was gonna, you know, I, [00:17:00] it was probably gonna be a struggle, chapter six. I was kind of like, oh, that's kind of like, you know, the. The, the little dessert, um, after it, but, but I found it harder than chapter five even. Yeah, me
Benjamin James Kuper-Smith: too,
Koen Frolichs: I guess me
Benjamin James Kuper-Smith: know a lot actually. Yeah,
Koen Frolichs: yeah, yeah.
Benjamin James Kuper-Smith: I mean, I have,
Koen Frolichs: and
Benjamin James Kuper-Smith: yeah, yeah. No, I mean, I, yeah, I mean, this one is also then just half the length. Uh, so here I basically only have two points I guess, to make, which are not exactly critical, but. Not exactly positive either. Okay. Um, one is that, uh, so I had this thought before when, so he talks about these, what they called logical positivists, whatever.
Mm-hmm. People who use the, the sim, um, symbolic level and then apply, you know, logical reasoning to it and how things work.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: And it just always seems to me like people wasting their time on something that's obviously not working. It just seems so silly. I dunno. You know what I mean? Like, it just seems like people just [00:18:00] putting so much effort and thought into framework that just doesn't seem to be going in the right direction.
Koen Frolichs: Yeah, yeah. Like, quit, quit, quit now guys. Yeah. Yeah. Before you get into deep. Yeah. Could be. Yeah. I, I, I feel like I'm not, I'm qualified to judge, judge these
Benjamin James Kuper-Smith: people. I mean, I guess I'm not, I'm not qualified either, but, but I guess the point is more that trying to. Use strict logical rules to explain at the symbolic level how people think.
Just seems slightly. Yeah. No, you're right.
Koen Frolichs: Yeah. Yeah, you're
Benjamin James Kuper-Smith: right. Like at the very best, it's a very incomplete picture of the whole thing, it seems to me.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: And yeah, to some extent, you know, I mean, I haven't thought this. You know, three obviously. I mean, I had something similar in the beginning when I heard about the whole Homo Economicals in the kind of economics rationality research where people, you know, psychologists criticize economists because they create this supposedly rational man and then say like, you know, oh, if this and this is the case, then people behave according to these rationality rules and that kinda stuff.
And I've now [00:19:00] like realized actually there is a lot of point to that. If you are aware of the kind of limitations of it.
Koen Frolichs: Sure.
Benjamin James Kuper-Smith: I'm not entirely sure where that applies here, but yeah. I dunno. I just kept having that thought when, not only this chapter, but several times of the book.
Koen Frolichs: It's symbolic people.
Yeah.
Benjamin James Kuper-Smith: Yeah.
Koen Frolichs: Yeah. It could be. Yeah. I, I, I, I. I mean, for me, I, I then wonder like, or is it me just not understanding it? Right. Like, do I not understand the, the depth of their reasoning? Maybe sometimes. I dunno.
Benjamin James Kuper-Smith: Yeah. Yeah. Okay. So that was, that was one point. And the second point is, so, so, you know, he talks quite a bit about the kind of analogy between humans and scientists in this one.
So how, like humans, humans reason about things in general and how scientists do it.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: And what I find that stuff kind of interesting. I'm always slightly. Hesitant to accept when scientists think that people act like scientists. I'm not entirely sure what my problem here is, but I had it the same way when, when I was studying predictive [00:20:00] coding for my master's thesis, where they basically said like, Hey, like perception doesn't work in this passive way, but instead people kind of create hypotheses and they test them through perception.
Koen Frolichs: Yeah,
Benjamin James Kuper-Smith: yeah. Um, and so I was like, yeah, you're just, yeah. I don't know exactly what it is, but like whenever a scientific theory matches up too closely to what the person is doing in their day-to-day life, it just seems like not the correct theory.
Koen Frolichs: There's this theory like, um, 'cause I felt like that's a little bit more of an older Id even, right, like earlier psychology, right?
Mm-hmm. Is that, that humans are rational also, that, you know, humans are more rational than. We think they Right. That is
Benjamin James Kuper-Smith: older.
Koen Frolichs: Yeah. Like, not like this cognitive miser theory then kind of came by Fisca and something, someone, um, I don't know, like kind of that humans kind of like more like, you know, like kind of based on the work by, by, um, Kaman and Ky like, you know, humans use heuristics way more.
Mm-hmm. Um, stuff like that, which I [00:21:00] think is, you know, intuitively seems more, um, you know, from my own day-to-day, um, heuristics, um, uh. Use? No, it just seems like a more, I mean, I guess not really what you're talking about or like it's little bit, it's more
Benjamin James Kuper-Smith: related to the previous point.
Koen Frolichs: Yeah.
Benjamin James Kuper-Smith: It seems to me, yeah.
Koen Frolichs: My, the, the book drew me off. I believe. It's fine.
Benjamin James Kuper-Smith: Yeah. I mean, we said this was gonna be a short one. It was already almost 25 minutes, so, yeah. Yeah. We talk too
Koen Frolichs: much.
Benjamin James Kuper-Smith: We, we, we have apparently more to say than we thought we did, or we talk more than we have to say. I'm not entirely sure which one
Koen Frolichs: I take it.
Benjamin James Kuper-Smith: Probably both.
Koen Frolichs: It's definitely, uh, too bad.
Benjamin James Kuper-Smith: Yeah. One of them. Definitely. The other maybe. Yeah, exactly. I'm not gonna say which is which.
Koen Frolichs: Yeah. Yeah. Yeah, no. Yeah, I think, I mean, and I have to admit, I think for chapter that that's also it for chapter six for me because I mean that the one thing is where he talks about this artificial neural networks, how they kind of, like, I kind of got lost there as well.
I don't think these [00:22:00] figures are very helpful. Um,
Benjamin James Kuper-Smith: I dunno. I mean, it's, I, yeah, I'm just not sure he knew exactly what he was doing. Yeah, that was I think, more the problem. I think the figures are probably useful once they're put properly into context. Um, but I agree. This was something I found slightly.
Yeah. I just, I mean, I read through it and at some point it's been like, yeah. I'm not sure I need to really go over this again so I understand it properly.
Koen Frolichs: Um, mm-hmm.
Benjamin James Kuper-Smith: Yeah, yeah,
Koen Frolichs: yeah, yeah, yeah.
Benjamin James Kuper-Smith: May maybe it makes more sense if you look at it properly. Um, but I kind of read through it and thought like, I don't know, what is he doing?
Uh, whatever.
Koen Frolichs: Yeah. Yeah. Okay. Yeah. Same. Same.
Benjamin James Kuper-Smith: Yeah. Okay.
Koen Frolichs: So one, one quick look into the future. What do we have?
Benjamin James Kuper-Smith: Right. Well, we have the rest now, which is 30
Koen Frolichs: pages in total, I think. Right? 33
Benjamin James Kuper-Smith: 2. 62. So it's chapter seven and eight, which are about computational aspects and in chase of space.
Koen Frolichs: Oh, that's a good title.
Benjamin James Kuper-Smith: Uh, which I guess is kind of [00:23:00] his summary. So I guess next time is gonna be those two. It's, yeah, it's said it's only 30 pages. Exactly. Uh, and I guess we're gonna have it. Maybe a bit of a general discussion also.
Koen Frolichs: Yeah, sure.
Benjamin James Kuper-Smith: And kind of see what we took from the book, I guess.
Koen Frolichs: Yeah. No, that makes sense.
Benjamin James Kuper-Smith: So I guess that's it.
TT YL. See again next week, like record is doing a P sign. For those who can't see on this audio podcast that's doing it, you should just,
Koen Frolichs: yeah.